AGARD-R-7 16

4J

AD-A141 456AGARD REPORT No. 716

Composite Structure Repair

~~A~ 1AM48 4This document 1has been approved'"6for Public r,-eQ3e and Sale; itsrdistribution is unlimited.

DISTRIBUTION AND AVAILABILITYON SACK COVER

AGARD-R-7 16

NORTH ATLANTIC TREATY ORGANIZATION

ADVISORY GROUP FOR AEROSPACE RESEARCH AND DEVELOPMENT

(ORGANISATION DU TRAITE DE L'ATLANTIQUE NORD)

AGARD Report No.716

COMPOSITE STRUCTURE REPAIR

by

Larry G.KellyAFWAL/FIBC

Wright Patterson AFBOH 45433

USA

Paper present, i at the 57th Meeting of the Structures and Materials Panel inVimeiro, Portugal on 9-14 October 1983.

THE MISSION OF AGARD

The mission of AG ARD is to bring together the leading personalities of the NATO nations in the fields of scienceand technology relating to aerospace for the tollowing purposes-

- Exchanging of scientific and technical information;

- Continuously stimulating advances in the aerospace sciences relevant to strengthening the common defenceposture;

- Improving the co-operation among member nations in aerospace research aaid dev..)pment;

- Providing scientific and technical a6vice and assistance to the North Atlantic Military Committee in the fieldof aerospace research and development;

- Rendering scientific and technical azsistance, as req'iested, to other NATO bodies and to member nations inconnection with research and development problems in the aerospace field;

- Providing assistance to member nations for the purpose of increasing their scientific and technical potential;

- Recommending effecti;',- ways for the member nations to use their research and development capabilities forthe common benefit of the NATO community.

The highest authority within AGARD is the National Delegates Board consisting of officially appointed seniorrepresenta tives from each member nation. The mission of AGARD is carried out through the Pane's which arecomposed of experts appointed by the National Delegates, the Consultant and Exchange Programme and the AerospaceApplications Studies Programme. The results of AGARD work are reported to the member nations and the NATOAuthorities through the AGARD series of publications of which this is one.

Participation in AGARD activities is by invitation only and is normally limited to citizens of the NATO nations.

The content of this publication has been reproduceddirectly from material supplied by AGARD or tht author.

Published Feb uary 1984

Copyright ACARD 1984All Rights Reserved

ISBN 92-835-1466-1

bt i

Printed by Specialised Printing Services Limited40 Chigwell Lane, Loughton, Essex IGIO 3TZ

-- y-ii

PREFACE

In preparation for a Specialist Meeting being planned for spring 1986, a pilot paper onthe subject "Composite Structure Repair" was provided to "The Repair of Aircraft StructuresInvolving Composite Materials" Sub-Committee at the 57th meeting of the AGARD Structuresand Materials Panel. Mr I arry Kelly, USAF, Air Force Wright Aeronautical Laboratories,presented in the pilot psper a summary of USAF experience in repairing in-service aircraftstructural composites. This paper has assisted the pane! in defining the context which shouldbe emphasized in the Specialist Meeting and the Sub-Committee is grateful for Mr Kelly'sassistance.

KEITH I.COLLIERChairman, Sub-Committee onThe Repair of Aircraft StructuresInvolving Composite Materials

IIJ1,

______1

iti

CONTENTS

Page

ABSTRACT

INTRODUCTION 1

U.S. ADVANCED COMPOSITE REPAIR EXPERIENCE

DAMAGE ASSESS A4ENT 2

BONDED REPAIRS 6

SUMMARY 17

REFERENCES 18

\

-it.

- -- -iv

Composite Structure RepairLarry G. Kelly

AFWAL/FI6CWright-Patterson AFB

Ab Lroct:" The technology for advanced composite structure repair is presently in adeveloping stage. The boundaries avid limitations of bolted versus bondedrepairs and precured patches versus cocured in place patches 3nd theirapplicability to various types of hardware has yet to be clearly established,.This paper does not discuss step by step repair procedures for specific air-craft components, such as defined in repair technical crders, but ratherprovides general guidelines for repair concepts and discusses two repairconfigurations that are generic in nature; an external patch and a nearflush repair and the extent to which they have been verified in the US.These repairs are applicable to a wide variety of light to moderately bonded(up to 25,000 lb/inch) stiffened and honeycomb sandwich structure sustainingdamage over a reasonably large area (up to 100 sq. in.). Also provided arereferences to documents containing step by step procedures for these repairtechniques and identification of organizations in the US. actively engagedin advinced composite structure repairz

Introduc tion:

major airfraie conuFenU, I, uu i, o auvanceu compos II te "-ia L e^(1 aa represently flying on a number of military production air,'raft in the UoS.Use, ini the Air Force, began with the F-15 and F-16 aircraft which employ1.6% and 2.5% advanced composites by structural weight. The Navy's F-18 andAV8B aircraft extended the use to 9.5% and 26% respectively, The Army ispresently evaluating a composite rear fuselage for the Black Hawk UH60 heli-copter which would extend the amount of composite structure utilized from 17%to 26%. This includes fiberglass, Kevlar and carbon materials, The Army isalso developing a composite helicopter prototype under the Advanced CompositeAirframe Program (ACAP) that will utilize composites for 75-80% of the air-frame by weight.

Until recently, advanced composite parts subject to major damage, werereturned to the manufacturer for repair. This situation is rapidly changingfor all three services are preparing to maintain aircraft that make extensiveuse of composite materials. Advanced composites are now being considered, inthe U.S. aircraft industry, for all aircraft structure applications where sub-stantial weight savings, stiffness or design efficiency requires tailoring thestructure for anisotropic load requirements.

U.S. Advanced Composite Repair Experience:

The service experience with advanced composites has been geoierally goodwith the exception of a few parts. Maintenance problems, for the most part,have consisted of edge damage or punctures and dents on composite coveredhoneycomb. These have been readily repaired by both field and depot levelpersorniel. These repairs have been generally non structural, that is, pe '-formed to prevent damage growth, provide aerodynamic smoothness or prevent

2moisture intrusion. The bounds for such repairs have been adequately definedby appropriate technical orders,, Several military repair centers are rapidlydeveloping the capability to do much more extensive repair and even majorcomposite structure remantfacture if necessary. Some of the more noteworthyfacilities in this regard are:

Naval Air Rework Facility - North Island, San Diego, CaliforniaNaval Air Rework Facility - Cherry Point, Ne-th CarolinaAir Force Logistics Center - Warner Robins AFB, GeorgiaAir Force Logistics Center - Hill AFB, Ogden, UtahAir Force Logistics Center - McClellan AFB*, Sacramento, CaliforniaArmy Depot - Corpus Christi, Texas

These facilities are being supported by several Research and Development organi-zations with background experience in advanced composites. The following R&Dorganizations are actively involved in composite repair technique development:

Naval Air Development Center - Warminster, PennsylvaniaNaval Research Labs - Washington, DCAir Force Wright Aeronautical Labs - Dayton, OhioArmy Applied Technology Lab - Fort Eustis, Virginia

Damage Assessment:

General impact damage and specifically ballistic penetration of a compositelaminate results in holes in the laminate which are irregular in contour andgenerally jagged in appearance. Delaminations,, void areas and rupturedfilament bonds may occur anywhere throughout the thickness, but generally toa larger extent on the opposite side of the impacted face or exit side of theprojectile path. In some cases, impacts which cause very little damage on thesurface can cause internal cracking and delamination. These interlaminardefects can be readily detected with ultrasonic equipment ind it is a goodrule that any damage which is visible on the surface ;hould be further evalu-ated for internal damage. Examples of extensive internal damage where surfacedamage is minor are shown in Figures 1 and 2.

*This facility was recently designated by the Air Force to be its leadcenter for establishing cormiposite repair training requirements for ALr.engineers and maintenance personnel. The center will develop compositerepair techniques including training and equipment needs, be a focal pointfor overall composite repair technology and 3id the other ALCs in implementingcomposite repairs.

__

IMPACT 1

S -

~

.

...-

- +. -, -o

'

%00 -

0.02 INCH i --

8 PLIES GR/EP (-: 4510/90)s

BLUNT IMPACTOR AT CENTER OF 5 INCH SQUARE AREATOTAL ABSORBED ENERGY 124 FT-LB(INCIPIENT DAMAGE INDICATED AT 0.82 FT-LB)DAMAGE NOT VISIBLE ON IMPACTED SURFACESLIGHT MATRIX CRACK ON BACK FACE

I .... 1 1*1hoo,,! r oqraph of Imnnr1" n %manp I aminate

Face Sheet: 4 Ply HMF 133 Impacted by 5/8 inch diameterWoven GR/EP Spherical Steel Impactor

Surface Indentation 0.021 inch Impact Energy 1.78 Ft-Lb.(No cracks or broken fibers) Impact

FM 300 Adhesive

Core: 6.1 PCF Al Honeycomb

Figure 2. Photomicrograph of Impact Damaged Honeycomb Panel

The primary field and depot inspection methods hding utilized in the U.S.for composit2 structure are through transmission and resonance ultrasonicsand radiography. Radiography inspection is used to detect broken hondlines(core splice and core to closeout members) and to detct the presence ofwater in the core cells. It can also be used to detect porous or excessivelythick bondlines and deformed core.

Ultrasonic equipment is the most widely used ano is generally employedwith a set of standa.ds for set up and defect compar'son. Figure 3 comparessize of visible dlmage to area of internal delaminations as determined byultrasonics. This data is from Reference 1 and is for a wide range of carbonpanel types; some with buffer strips and stitching to contain delayed andsuperquick fuzed 23 mm high explosive projectile damage. The original datais from MIcDonnell Aircraft but I have included data from Boeing, Northropand Air Force reports (dots and bars). This data includes impacts of frag-ments (1/4, 3/8, 1/2 inch) and projectiles (12.7, 14.5, 23 mm) with angles ofobliquity up to 60 degrees.

16 -0.20 AS113502 MONOLO'HIC PANEL23 MM HEI BALLISTIC DAMAGE A5

I E212 B2

UPPER El1c A4DELAMINATION BOUND-\jDAMAGE _ _

IN.SIMULATED BALLISIIC DELAMINATION

ENELP 8E 2,A ADMGDAMAGE FROM B131 DAMAGEPENETRATR - VISIBLE

goo

Al1

;1 000 2 4 6 8 10 12

VISIBLE DAMAGE ENVELOPE - IN.

Figure 3. Ballistic Damage to Carbon Epoxy Panels

Some test results, Reference 2, indicate that for a given panel widthand laminate orientation various through-the-thickness crack geometries havingthe same crack width, as shown in Figure 4, failed at essentially the sametension load. Thus assuming damage to consist of a through-the-thicknessdefect equal in width to the maximum damage dimension (as determined by ultra-sonics) perpendicular to the primary load path, linear elastic fracture

I.

mechanics can be utilized to obtain an estimate of the strength lost. Thisapproach of modeling damage effective strain concentrations as that of anequivalent open round hole can sometimes be unconservative but a usefultechnique to obtain a "ball park" estimate of how much strength his been lost.

A W_ .,W.---W -- -,,1.d W -- l ,. W -' ,b .- W .--I

2a a 2.2 j: . 2a .LA. HORIZONTAL B. CIRCULAR C. SQUARE D. ANGLE E. IMPACT

SLIT HOLE HOLE SLIT DAMAGE

Figure 4. Examples of Through-Thickness Defects Having Same Tension Failure Loads

Figure 5, from Reference 3, shows that there is substantial strength loss incarbon composites with relatively small holes.

HOLE DIAMETER, INCH

0 0.5 U.0

(5 I nzI

00. AS/3501-5 [o,,1/o2/A,1. ,< O0- 0 t300/5208 [0/45/902,bZ ( 1300/934 L(0/45/0/90),1 2

U MOD II1206 [0/*45/90],80 - 0 T300/934 [(0/t45/0/90),]2

00-60

"~~ 0.5 0. 1, .0253 .

40

20

REGION AT END OF SLOT

0 0.5 1.0 1.5 2.0 2.5 3.0 3.5HOLE DIAMETER, cm

Figure 5. Tensile Strength Retention of Laminates with a Hole

-_- -- . -U - --- - .. . .

Current design practice in the U.S. is to liit ultimate strain al lowadhliesin ca rbon compos i te structures to app rox i ma te 1 y 3 -00 -S n /09. I n Tb i s allow(sfor stress concentrati oins due to 0ol t holes orf notches and provides for sofieaccommodation o0 st rain COtiet re. i ions dlu to deeccts or daiiiage. The I i gure5 data does, however, point out (he need Iotr good tladlily repali's with sUb-stantial load cIarrying capabilit y C.specially tor structuros designed withhigher strain allowables.

BondedN epai rs

Two types of bonded repairs are discussed below: 1) a nearly flsh repliirfor which a scarf joint surface iS Machined in the pairoent laminate and replace-inent plies with l adhesive ore cocured into place; 2) an external patch which isprecured and subsequently bonded over the damaged area. These repairs can beused for on aircraft or off aircraft repairs, for repairs accessihle only fromone side for either flush or external patches arid for both iionolithic or sand-wich construction. The information provided is not intended to be a step bystep guide for repair patch installation such a; found in References 4 and 5,but rather a discussion of standardized repair proceoures that have been veri-fied, and general engineering guidance for the designer of the repair patch.

After assessing the damaqe and before decidingy Upor; a repair, the questionof the parent laminate moisture condition becomes important. Moisture absorbedin the laminate and/or entrapped moisture in honeycomb can be very detrimentalto the integrity of bonded repairs. I xari nation of cured carbon epoxy patchIesi)onded to substrates containing moisture, similar to long term service experiencein a hi gh moisture environmen ', showed a porous bond line. See Figure 6:

- flnSu n- x ' r - o.

AA

Figure 6. Porosity in AI-147 Bondline on 50-Ply Wet Laminate

This absorbed moisture has had detrimental effects on repairs in thefollowing four ways:LL

71) Local delamination or blistering in parent laminates2) Reduced strength of the repair and repair bond line resulting

from purosity.3) Expanding moisture in honeycomb cells has created sufficient pressure

to separate the skin from the core.4) Reduced effectiveness of ultrasonic inspection due to strong signal

attenuation making it difficult to verify bond line integrity.Prebond drying (a minimum of 48 hours at 17oF-2OOF), slow heat up rates,

reduced cure temperatures and selection of adhesives lIss sensitive to moisturecan minimize or eliminate the above problems. The 250 F curing adhesives, as agro8 p, are more sensitive to prebond moisture at higher temperatures (above150 F) than 350 F curing adhesives. Drying the parent laminate to an averagemoisture content of less than .5 percent is recommended. This can be verytime consuming taking over 24 hours for a 16 ply laminate, as shown in Kigure 7.

1.__0_ _ _ _ _

0.8 DRYING TEMPERATURES, :,75 OF

PERCENT 0.6 'MOISTURE ~ tl

CONTENT

0.4 i.. .

0.2

0 24 "1 250)F

0 ! 24 HR

0 100 200 300 400 500

THICKNESS IN.

Figure 7. Drying Time for Carbon Epoxy Laminu,.es

External bonded doublers are the simplest to apply. Their load carryingcapability is, however, some hat limited for no matter how well the edge ofthe patch is tapered the edge of the parent laminate, at the hole, is a pointof high shear and peel stress concentration. Smnce the interlaminar tensilestrength of carbon epoxy is less than the peel strength of typical structural

adhesives, the effect cf this failure mode is to restrict tne thickness ofcomposites which can bc bonded efficiently using star dard lap joints,Reference 6 shows that peeling can be minimized by small fasteners at aboutone inch spacing around the hole and a router cut about o017 inch deep filledwith adhesive and a ply of fiberglass prepreg. This concept shown in Figure8. on a 16 ply [(+45/0/90) laminate using 1/8" blind rivets raised thejoint efficiency Trom 52% t0 78%.

FM-400BONDED JOINTS FAILURE JOINT

(WIDTH - 1.00 INCH) MODE EFF

PEEL AND SHEAR"-m-FZI.,IIZZI1YFAILURE

CONTROL

SHEAR AND PATCH NET 0.73"TENSION FAILURE

ONE RIVET

.F: _e SHEAR AND RIVETHEAD PULL-THRU 0.78

FAILUREUNDERCUT & ONE RIVET

Fi'-ure 8. External Patch Concepts

Applying this concept to 2" diameter holes in four-point-load sandwichbeams obtained tha results shown in Table 1. Figure 9 shows the details ofthe 22 ply patch design. Comparing the resulting failure loads with the pa.'entallowab.es quite satisfactory results were obtained for low to intermediateload levels (6-i,000 Ib/inch)

PRECURED PATCHAS/35014, 22 PLIES I /8.IN DIA CSKBLIND RIVET

CURED CONTOUR

-00.25 3-. ,68

FMAO ., It45) GLASS/EPOXY